When New Ventures Meet Technology: What are the mechanisms through which technology influences the structuring process of new ventures?

(1)

When New Ventures Meet Technology: What are the mechanisms through which technology influences the structuring process of new ventures?

Kay Haze (S2585766)

March 21, 2016 Word count: 49.917

Master Thesis Business Administration: Specialization Change Management Faculty of Economics and Business

University of Groningen

(2)

2

Abstract

This research investigates the role of enterprise systems in the organizing process of new ventures. Past research primarily focussed on the role of technology in established structures. Since new ventures operate outside established structures and mostly start with a “blank slate”, technology could play a different role in their organizing process. In this study the methodological principles for conducting critical realist case studies by Wynn and Williams (2012) were applied, and four new ventures were investigated. Through retroduction, two mechanisms emerged contributing to the organizing process of two organizations in this study: An embeddedness and a conditioning mechanism. These mechanisms resulted in copying of the embedded organizational elements from the system by the entrepreneurs. In doing so the entrepreneurs were releaved of the burden of having to develop certain routines and templates of their own, and their ventures organized faster.

(3)

3

Acknowledgements

(4)

4

List of tables

Tab. 1: Data collection overview

List of figures

Fig. 1: Potential relationships between materiality, sociomateriality, and socio-technical systems (Leonardi, 2012, p. 43).

Fig. 2: Relationships of CR Methodological Principles (Wynn and Williams, 2012, p. 797) Fig. 3: A critical realist model on technology-mediated organizational change (Volkoff et al.,

2007, p.843)

Fig. 4: Evaluating Causal Explanations (derived from Runde, 1998) (Wynn and Williams, 2012, p. 802)

Fig. 5: Timeline [Company_A] Fig. 6: Timeline [Company_B] Fig. 7: Timeline [Company_C] Fig. 8: Timeline [Company_D]

List of acronyms

CLA Collective labour agreement

CR Critical realism

CRM Customer relationship management ERP Enterprise resource planning

ES Enterprise system

FTE Full-time Employee

IS Information system

IT Information technology

SCM Supply chain management

(7)

7

1. Introduction

“Our steadfast belief is that IT is the magic ingredient that inspires and most often enables contemporary entrepreneurial endeavours” – (Giudice and Straub, 2011)

During the past 15 years, IT has played a central role in the entrepreneurial development of virtually every country in the world, as well as contributing crucially to the performance of advanced economies (Giudice and Straub, 2011). Whilst in the beginning of the past century an architect would be able to set up a business with merely his education, paper and a pencil, todays entrepreneurs are not only affected by the rules and regulations imposed on them by governments and institutions but are also often, but not always, bound to use certain information technology in order to run their business.

When examining how entrepreneurs, nowadays, organize their new ventures, literature shows that the routines, habits and heuristics formerly acquired by entrepreneurs lay a first foundation for developing a certain structure within their organization (Aldrich and Yang, 2013). In further organizing their venture, entrepreneur can benefit from actively working on the organization (learning by doing) and through experimentation, either deliberate or accidental via trial and error (Aldrich and Yang, 2013). However, rather than adopting a trial and error approach, many entrepreneurs will attempt to follow or imitate templates, when faced with the challenges of developing new routines (Aldrich and Yang, 2013).

Overall, a new venture unfolds as a structure is created and modified, resources are acquired and combined, a culture takes root, processes and routines are implemented and updated or replaced, and core elements of a business model are integrated in different ways until something emerges that works (Morris and Webb, 2015). A process of organizing is initiated by the entrepreneur wherein organizational components (e.g. people, information, technology) combine in (un)predictable ways and the component parts form something unlike the parts themselves (Morris and Webb, 2015). This interaction among components leads to the creation of novel, resultant emergent states (McKelvey, 2004). Moreover, “Unlike

managers in established organizations, who generally follow or modify pre-existing routines already selected by others, entrepreneurs begin with mostly a blank slate” (Aldrich and Yang,

2013, p. 60).

(8)

8 technology (e.g. Orlikowski, 1992, 2000; Volkoff et al. 2007). Specifically focussing on enterprise systems, Volkoff et al. (2007) found that organizational elements such as routines, roles and data structures could become embedded within the enterprise system, giving them a material aspect. Being embedded, these elements mediated human action, resulting in a change of the organization’s structure under study.

Acknowledging that new ventures start with mostly a blank slate, and will likely imitate certain templates when faced with the challenges of developing new routines (Aldrich and Yang, 2013), it becomes interesting to see what role technology plays within the organizing process of these new organizations. Especially, what the mechanisms are through which technology affects the structure of new ventures.

No research has yet attempted to explicate these mechanisms within the particular context of a new venture. Hence, in this study an attempt is made in retroducing the mechanisms most likely contributing to the organizing process within four new ventures, with a specific focus on the enterprise systems introduced within these organizations. Doing so will contribute to the field of information systems studies by offering an explanation to how technology affects the structure of new ventures. Moreover, entrepreneurs will benefit from knowing these mechanisms, increasing their understanding of the possible effects and benefits technology can have on their organization’s structure. Overall, introducing technology within a new venture will result in adding yet another ingredient to the soup being the organization. Understanding its effects will help future entrepreneurs comprehend the consequences and benefits when introducing technology within their organization, and how important it is to combine the correct ingredients for the organization one wishes to create.

To answer the questions emerging from the research gap identified above, this study makes use of the critical realist principles for conducting case study research by Wynn and Williams (2012). Moreover, this study elaborates on the study by Volkoff et al. (2007) and adopts a critical realist view in which the social and the material are viewed as conceptually distinct. This allows to logically discussing their effects on each other, and acknowledges the fact that all social structure logically predates the actions it engenders (Volkoff et al., 2007). Put more simply, technology, although being socially shaped, can exist as a separate entity before being put to use within an organization.

In short, the following research question is applied within this study:

(9)

9 In order to answer this question, case studies were conducted within four separate new ventures. Most critical realist researchers have identified the case study method as the best approach to explore the interaction of structure, events, actions, and context to identify and explicate causal mechanisms (Ackroyd, 2010; Easton and Harrison, 2004; Miles and Huberman, 1994). Moreover a case study allows gathering a rich amount of data, which helps to untangle the critical events and structural forces at play within these organizations.

Through semi-structured interviews, informal conversations and observations within these four organization data was gathered. The data was processed in Atlas.ti v7.5 and coded using open, axial and selective coding (Strauss and Corbin, 1998).

In the next chapter, the concept of sociomateriality and new ventures are introduced, in order to gain an initial understanding of these main concepts. Moreover the model by volkoff

et al. (2007) serves as an example of what we already know about the role of technology

(10)

10

2. Conceptual foundations

2.1 Materiality and Organizations: The case of Enterprise Systems

Although the quotation at the beginning of this research paper might suggest that technology has just recently become an important factor for enabling entrepreneurial endeavours, it is important to realize that both humans and technology have a long history of mutually shaping each other. For as long as humans have been around, we have created, shaped and build certain “technologies” aiding us in realizing our goals. A simple rock shaped in the form of a knife enabled us to cut certain things, and writing, also being socially shaped, allowed us to communicate in an entirely new way as opposed to the vocal communication we used before. In essence many innovations have been at the foundation of enabling entrepreneurial endeavours, and both people and technology have been shaping each other for decades. It is via these former illustrations that we can comprehend the fact that all materiality is social in that it was created through social processes and that all social action is possible because of some materiality (Leonardi, 2012).

With time both our goals and technologies evolved, as did the words we use to define these concepts. Researchers nowadays prefer using the word sociomateriality instead to highlight the important interplay between these two concepts. Materiality is defined as “the

arrangement of an artifact’s physical and/or digital materials into particular forms that endure across differences in place and time and are important to users” (Leonardi, 2012, p.

31). When speaking of the social we primarily insinuate human action, or in the case of organizations, a group’s coordinated exercise of forming and realizing its goals (Pickering, 2001)

When relating these concepts to organizations, and the technology these organizations use, another important separation is that between both human and material agency. Material agency is typically defined as the capacity for nonhuman entities to act absent sustained human interventions (Leonardi, 2012), whilst human agency is defined as the ability to realise one’s goals (Giddens, 1984). For clarification, “whereas materiality refers to properties of the

object, material agency refers to the ways in which the object acts when human provoke it”

(Leonardi, 2013, p. 70).

(11)

11 subsystem of an organization, which is comprised of abstract social structures such as roles, statuses, hierarchies and other similar constructs also referred to as institutional forces (Kallinikos, 2011). Conversely, the social subsystem affects sociomaterial practice, enabling and constraining certain actions (Leonardi, 2012). In short, it is through the interplay of sociomaterial practice and the social subsystem, that an organizations structure is gradually formed. This structure is the product of past imbrication processes, a result of the gradual overlapping and interlocking of distinct elements into a durable infrastructure (Leonardi, 2013). Figure 1 gives an example of how this process unfolds.

Figure 1. Potential relationships between materiality, sociomateriality, and socio-technical systems (Leonardi, 2012, p. 43)

(12)

12 system could, for example, be an enterprise resource planning system (ERP), a customer relationship management system (CRM) or a supply chain management system (SCM).

As is the case with materiality in general, enterprise systems can, given the goals one wishes to realize through it, both afford and constrain one in realizing their goals (Leonardi, 2011). Secondly, it’s because of the beliefs one has about what technology allows one to do, how and if the technology is used in realizing one’s goals (Leonardi, 2013).

A toolbox for example offers a wide arrange of tools, which can enable one in realising their goals. Would their goal be to drive a nail into a piece of wood, the obvious choice would be to select the hammer for this job, since it was specifically designed for enabling one to realize that goal. Essentially, one’s field of reference or former experience with this activity urges them to select the hammer in this situation, and not the other tools since they serve yet another purpose. Would the situation occur that the person in question would misplace a nail, the back end of the hammer would enable him or her to remove the nail in that given situation. Although a hammer consists of both a front and a back end, the context plays an important role in which material aspects of the ‘technology’ are being used. Thus, although technologies can contain multiple material aspects both the context and ones beliefs about its usefulness, determine whether these aspects are used. Once appropriated, this use results in mediated action by the technology.

With respect to enterprise systems, past research has already examined the effects of these systems within organizations, with a specific focus on looking at what changes are caused by its introduction. In particular the study by Volkoff et al. (2007), using a critical realist base, identified the mechanisms causing the changes experienced within an organization introducing a new enterprise system. In essence, routines, roles and data structures had become embedded within the technology, giving these a material aspect (Volkoff et al., 2007). They had become fixed in a certain way, thus becoming of influence to its users.

(13)

13 pressing of a button, for it to be able to work. In essence the automatic screwdriver has embedded a certain routine, since it now automatically does “all the work.”

Turning back to enterprise systems, Volkoff et al. (2007), primarily using routines as a unit of interest (Nelson and Winter, 1982), identified materiality to mediate between the ostensive and performative aspects of a routine, bringing these closer together. The ostensive aspects of a routine are the routines in principle, whilst the performative aspects of a routine are the routines in practice (Feldman and Pentland, 2003). In the case of enterprise systems,

“organizational routines are embedded in the ES in the form of system-executed transactions sets of explicitly defined steps that require specific data inputs to automatically generate specific outcomes” (Volkoff et al., 2007, p. 839). It is because of the fact that users were

confronted with technology which had already been preconfigured for them that their routines became mediated and eventually an organization was changed due to imbrication of social and (new) material agency in the organization its structure (Leonardi, 2013; Volkoff et al., 2007). Although technologies can eventually be changed, even the perceptions of what changes could be made are constricted, to a large degree, by the initial materiality users’ encounter when first using the technology (Leonardi, 2013).

Overall, the fact is that enterprise systems, when introduced within organizations, can cause changes to the organizations structure, because of its mediating effects on its users when attempting to realise their goals. The material aspects of the ES used will determine to what extend an organizations structure changes and which routines are mediated, the example of the toolbox helps to clarify this. Although the effects of introducing an enterprise system within established firms have been studied thoroughly, (recent articles by Volkoff et al. (2007), Mutch (2010), and Leonardi (2011) are good examples of this) no research has yet attempted to examine its effects on new ventures. Since new ventures are at the genesis of their ‘process of organizing’ the effects of materiality might be different than is the case in established firms, whose structure is already in place (Aldrich and Yang, 2013). The next section will focus on the concept of new ventures, how these entities come to be, and how these ventures differ from established organizations.

2.2 New ventures: organizing at its genesis

(14)

14 introducing technology will most likely result in changes to these structures, as was explained in the article by Volkoff et al. (2007). In contrast to established firms, new ventures are characterized as of not having any fixed set of institutionalized routines yet and are primarily formed through an entrepreneurs past experiences, complemented by a process of learning by doing, copying or borrowing, and experimentation (Aldrich and Yang, 2013).

The individual starting an organization is initiating a process of organizing. In essence a venture unfolds as a structure is created and modified, recourses are acquired and combined, a culture takes root, processes and routines are implemented and updated or replaced, and core elements of a business model are integrated in different ways until something emerges that works (Morris and Webb, 2015). Furthermore, Tsoukas and Chia (2002, p. 570) explain that “organization is an attempt to order the intrinsic flux of human action, to channel it

towards certain ends, to give it a particular shape, through generalizing and institutionalizing particular meanings and rules.” Thus, as was mentioned at the end of the previous section,

new ventures are at the beginning of their organizing process in which an organization’s structure is gradually given form through the interplay of human agency and the material aspects shaped by, or introduced in this process.

A key characteristic of the new venture creation process is emergence (Lichtenstein et

al., 2007). Emergence refers to order creation in which novel and coherent structures and

patterns are derived during self-organization (Goldstein, 1999). The entity is in the process of becoming something it was not before. Emergence is not simply a result of the process of interactions (i.e. a new state or entity), rather it takes place during the process of interacting (McKelvey, 2001). Emergent properties arise when components (e.g. people, information, technology) combine in (un)predictable ways and component parts form something unlike the parts themselves, this interaction among components lead to the creation of novel, resultant emergent states (McKelvey, 2004).

Overall, emergence focuses on the nature of interactions, including the properties of irreversibility, adaptive tension, nonlinear change (surprise/unpredictability) reciprocal interactions and supervenience (i.e. how the emergent order can, in turn, impose structure on its underlying components (Sawyer, 2001)).

(15)

15 interactions and their outcomes at the very least become a part of individuals’ memories that shape future perceptions, beliefs, and, ultimately, behaviour (Morris and Webb, 2015). The previous shows resemblance with a statement by Leonardi (2013, p. 69), who mentions that:

“although technologies can eventually be changed, even the perceptions of what changes could be made are constricted, to a large degree, by the initial materiality users’ encounter when first using the technology.”

Secondly, the property of supervenience refers to how the emergent order can, in turn, impose structure on its underlying components (Sawyer, 2001). In essence, during the process of organizing, the entrepreneur provides incentives and establishes controls that do not only prescribe how others employees should behave, but also, to some extent, how the entrepreneurs themselves should behave (Morris and Webb, 2015). When relating this to the concept of sociomateriality, the social subsystem of a new venture is shaped through sociomaterial practice. In essence, the ostensive aspects of a routine are gradually formed during the process of organizing.

In short, “Unlike managers in established organizations, who generally follow or

modify pre-existing routines already selected by others, entrepreneurs begin with mostly a blank slate” (Aldrich and Yang, 2013, p. 60). Moreover, by enacting their efforts outside

established structures, entrepreneurs cannot be subject to path dependency through bureaucratic mechanisms (McMullen and Shepherd, 2006). They do however encounter contexts in which other people, vendors, investors and regulators already have their own expectations concerning entrepreneurship (Aldrich and Yang, 2013). These expectations can both constrain and educate entrepreneurs as to what they could best do in particular contexts (Aldrich and Yang, 2013). Finally, although entrepreneurs may bring some knowledge of pre-existing routines with them, when faced with the challenges of developing new routines, many will attempt to follow or imitate templates that others have used, rather than adopting a trial and error approach (Aldrich and Yang, 2013).

(16)

16 process of becoming an organization, in which each entrepreneur pursues a unique path, with each experienced event informing and shaping the next (Morris and Webb, 2015).

Since new ventures start out with no institutionalized structure yet, and its routines are particularly shaped through a process of learning by doing and copying (Aldrich and Yang, 2013), an enterprise system could have a different impact when introduced within the venture as opposed to when it is introduced within an established firm. The system could, in theory, play an influencing role during a new venture’s organizing process, given the fact that it already embeds certain organizational elements such as role, routine and data structures. These embedded elements could form a source for entrepreneurs to draw upon when dealing with the difficulties of developing new routines. The following chapter will therefore dig deeper into the theory provided by Volkoff et al. (2007) in order to gain a better understanding of how this organizing process works out, and what role technology plays herein.

2.3 A Critical Realist Theory of Technology-Mediated Organizational Change

In essence both the concept of emergence and sociomateriality (from a critical realist perspective) show great resemblance and both adopt an open system perspective in which generative mechanisms form the ingredients for certain events to occur. Although both established organizations and new ventures differ with regards to the complexity of their organizational structure, in which the former already has an institutionalized structure in place and the latter begins with a mostly blank slate (Aldrich and Yang, 2013), both are alike in the sense that they are engaged in the process of organizing.

(17)

17 The first phase of the model, being structural conditioning, incorporates the pre-existing structural properties that are the consequence of past actions. Through sociomaterial practice an enterprise system is given shape and routine, role and data structures become embedded within the technology. As was mentioned before, “users are introduced to

technology whose materiality has already been preconfigured for them” (Leonardi, 2013, p.

69). In the case of new ventures these embedded elements can form a source for the entrepreneurs to draw upon when dealing with the difficulties of organizing a new venture. Moreover, simply by using the system to achieve certain goals, the entrepreneur will be conditioned by the initial materiality he or she will encounter, forming a foundation for future perceptions, beliefs, and, ultimately, behaviour (Morris and Webb, 2015).

The latter is likely to occur during the phase of social interaction, during which agents engage with, and are constrained and enabled by, the pre-existing structural conditions (Volkoff et al., 2007). It is here that human agency is channelled by the material aspects of the enterprise system they use when attempting to realize their goals. In the case of routines, the ostensive and performative elements of a routine are mediated by the material aspects used to perform the routine. This brings these closer together. The new venture, when using the enterprise systems, will be conditioned by the system regarding the way in which they should execute their tasks. This will eventually result in the final phase of the model of structural elaboration or reproduction, which is concerned with the modification of previous structural properties and the introduction of new ones or the reinforcing of existing structures (Volkoff et al., 2007). In essence, based on the feedback by its users, an enterprise system designer might choose to alter the material aspects to better enable its users to realize their goals. Secondly, when not changing the technology, the technology will stimulate reproduction of the imposed/current structure. New ventures are stimulated by the system to adopt the embedded organizational elements, which will then form the foundation for how certain routines will be executed in the future. Thus, although social interaction is continuous, at any moment the emergent structure depends on past activities, not on the actions of current agents (Volkoff et al., 2007). Users in turn are shaped and reshaped as they engage with the structures they confront, but did not create (Archer, 1998).

(18)

18 Moreover, “a perspective on sociomateriality based on the foundations of critical realism can

perhaps say more about the constitution of organization than can be said when analysed from an agential realist perspective, which maintains its focus on practice and simply assumes that an abstract notion of organization is merely the reflection of practice at a higher level of analysis” (Leonardi, 2013, p. 73). It will help answering the role materiality plays in the

(19)

19

3. Methodology

When using critical realism as a foundation for understanding the effects of enterprise systems on the organizing process of new ventures, it is necessary to adopt methodology that matches its underlying ontology and epistemology. Ontologically critical realism is based on the following basic assumptions: existence of an independent reality; a stratified ontology comprised of structures, mechanisms, events, and experiences; emergent powers dependent upon but not reducible to lower-level powers; an open systems perspective (Bhaskar, 1975, 1986, 1998; Colier, 1994; Danermark et al., 2002; Sayer, 1992). A key aspect of critical realism is stratification of reality into three nested domains (Bhaskar, 1975) “The domain of

the real includes the entities and structures of reality and the causal powers inherent to them as they independently exist. The domain of the actual is a subset of the real that includes the events that occur when the causal powers of structures and entities are enacted, regardless of whether or not these are observed by humans. The final domain, the empirical, is a subset of the actual and consists of those events which we are able to experience via perception or measurement” (Wynn and Williams, 2012, p. 790). Since mechanisms operate in an open

system, their effects may or may not occur, and those that occur may or may not be observed (Bhaskar, 1975; Outhwaite, 1998). The purpose of a critical realist study is to explain a given set of events by uncovering the hypothesized existence of mechanisms which, if they existed and were enacted, could have produced these events (Bhaskar, 1975; 1998). Secondly, a primary objective of scientific research conducted under critical realism is to develop explanations for the way things act and how they are capable of doing so (Wynn and Williams, 2012).

When investigating the mechanisms through which an enterprise system influences the organizing process of new ventures, the critical realist foundation suits well since it adopts an open system perspective and recognizes materiality as a separate entity (Volkoff et al., 2007). This allows investigating the role of materiality in the organizing process of new ventures.

Although various methods could be applied for answering the main research question (Zachariadis et al., 2013), most critical realist researchers have identified the case study method as the best approach to explore the interaction of structure, events, actions, and context to identify and explicate causal mechanisms (Ackroyd, 2010; Easton and Harrison, 2004; Miles and Huberman, 1994). A case study: “investigates one or a small number of

social entities or situations about which data are collected using multiple sources of data”

(20)

20 For case study research, a key concern is methodological triangulation (Wynn and Williams, 2012). Triangulation implies the concurrent, but separate collection and analysis of different, but complementary data on the same topic (Morse, 1991) Secondly, since critical realists seek to explain the why and how of a given complex event to occur, each having a unique context and structural influences, the emphasis should be on the detailed and precisely focussed study of a limited number of cases in a specific setting in an attempt to build explanatory theory that matches the empirical facts as closely as possible (Sayer, 1992). Therefore, a maximum number of four cases were conducted.

In order to structure the continuing of this research, the methodological principles by Wynn and Williams (2012) are employed to support the conduct and evaluation of CR case study research. The principles include explication of events, explication of structure and context, retroduction of mechanisms, empirical corroboration of hypothesized mechanisms, and adoption of triangulation and multiple research methods (see Figure 2.).

In short, the first step involves identifying and describing the critical events, which delineate the phenomenon of interest. Events are specific happenings resulting from causal mechanisms being enacted in some social and physical structure within a particular organizational context (Wynn and Williams, 2012). These events are commonly abstracted from many empirical experiences related to the outcome of interest (Williams and Karahanna, 2013). The second step, the explication of structure and context, builds from this previous analysis to identify what in the research setting is causally relevant, whereas it also provides a means for the description of key events. The explication of structure and context analyses the elements of social and physical structure as well as variations in contextual influences within the study setting to identify what is causally relevant to the identified events (Wynn and Williams, 2012). In short, the idea is to explain what about the structures, operating in that particular setting, might produce the events (Sayer, 1992). In general, these two first steps occur simultaneously and iteratively. Subsequent, the logic of retroduction is applied to describe potential causal mechanisms that could explain the observed events. Finally, the presence and efficacy of the mechanisms are confirmed by means of empirical corroboration (Williams and Karahanna, 2013).

(21)

21 In the continuing of this section the separate cases will be described as goes for the data collection methods and the data analysis strategy.

Figure 2. Relationships of CR Methodological Principles (Wynn and Williams, 2012, p. 797)

3.1 Data collection methods

To address the research question, cross- sectional case studies were conducted within four separate organizations. These organizations are characterized by the fact that they all meet the criteria of being a new venture and are no more than six years in business. Secondly, all of the organizations make use of an enterprise system, either custom build or acquired via an external vendor. Hence, although difficult from a critical realist viewpoint, minimum variation sampling is applied. Minimum variation sampling focuses on a gathering data from sources who share similar traits of specific characteristics (Yin, 2014). Although the organizations in this sample share similar traits, the context in which these organizations operate, the enterprise systems used, and the background of the owners all differ. This could result in different mechanisms emerging per case, resulting in different outcomes.

(22)

22 Complementary, observations and/or informal conversations were captured in field notes. In order to guarantee the accurateness of the data, after having transcribed the recordings, the documents were sent to the interviewees for confirmation. Table 1 Provides an overview of the data collected.

Table 1. Data collection overview

Interviews Fieldnotes Total

(Transcribed)

Source Semi-Structured

Interview

Observations Informal conversations

Company_A.1 30 Min. 30 Min. 40 Min.b 8 Pages

Company_B.1 27 Min. 20 Min.c 10 Min. 14 Pages

Company_B.2 32 Min.

Company_C.1 41 Min. 30 Min. 30 Min. 10 Pages

Company_D.1 44 Min. 30 Min. 20 Min. 12 Pages

3.2 Case & event descriptions

In this section a short description of the cases is provided. Due to similarities between the case descriptions and the explications of events, these sections were merged. A more elaborate case analysis section is however included in appendices B to F.

Company_A: The first organization developed a marketing tool, which allows their customers

to get an overview of their audience by use of Wi-Fi technology. Initially starting out with two persons, being the owner and co-owner, the organization has just incorporated a custom build CRM system aimed at supporting them during their administrative routines.

During the start-up phase of [Company_A] the entrepreneurs of the organization started a process of learning by doing. By running in to different situations the owners gradually developed an initial way of performing their routines. This resulted in the creation of several templates, in Word or Excel, used for storing customer data such as contact data, customer preferences, and contracts.

Due to frustrations regarding the amount of steps required to process a new customer, and the realization that the current way of working was ineffective the owners decided to design a system aiding them in performing their routines. The system was developed by an external party, although most input came from the owners themselves.

b_{[Owner_A.1] offered to meet informally up front, during this informal converstation the owner introduced his}

organization and the enterprise system used in the organization.

c

(23)

23 After introducing the system in the organization the number of steps required to process a new customer was reduced from twelve to three steps. Moreover, the owners are now able to see where, in the installation process, each customer is and who is responsible for the particular customer. In essence, introducing the system resulted in higher transparency, higher efficiency, a reduced chance on making errors, and automation of administrative processes. One interview was conducted with [Owner_A.1] one of the owners of the organization.

Company_B: The second organization is a development house, which focuses on the

development of communication (VOIP) and CRM systems. The organization is a split off of a larger holding and started with both a new name, and a new way of structuring roles within the venture. Being accustomed to using a CRM system called Sugar, the system formed a foundation for the continuing of the company, since it stored all data of current customers and companies. The system was primarily used next to other systems each serving a different purpose. Essentially the employees had to switch between different systems depending upon what they wanted to accomplish.

Since this former system was quite rigid, and the organization had acquired the necessary resources to make a system of their own, [Company_B] designed their own system, which incorporated all separate systems, enabling the employees to work from one system only. In short, the custom tailored system resulted in employees being able to work more efficient and data becoming more transparent within the organizations. Furthermore some routines previously performed manually by the employees of the organization had been automated by the system.

Two interviews were conducted with [Sales_B.1] & [Sales_B.2] of the sales department, both users of the CRM system.

Company_C: The third company is an employment agency, which has been in business for six

years. The [Owner_C.1] has previous managerial experience as head of another employment agency. [Owner_C.2] has previous experience in employment agencies as well, and has previous start-up experience.

(24)

24 reminded its users that some employees should be given another contract, this due to the collective labour agreement (CLA) used by the employment agency. In essence the users could only continue their normal activities after the employees were given a new contract. As such the system structured certain activities and played a more important role within the organization.

Only recently the organization started working with yet another system because the designer of the former system passed away. Although the old system allowed some creativity of its users, regarding for example the contracts of the temporal employees, the new system was stricter on how certain data should be inputted. This resulted in the employees working more precise and should in the long run result in less administrative tasks.

Overall, the system motivated its users to work more accurately. Moreover, it forced the owners to work differently since the former system was employee oriented, and the new system was organization oriented.

One interview was conducted with [Owner_C.1], both owner and founder of the employment agency.

Company_D: The fourth organization is an IT company, which focuses on selling thin clients

and other IT hardware and software. Being founded in 2011 the company has grown to a size of six FTE’s.

During the early phases of the organization [Owner_D.1] had some frictions with his companion who was responsible for the administrative tasks within the organization. After having bought out [Owner_D.2] the owner, with the aim of linking the webshop of the organization to a system, acquired an enterprise system called Afas.

Afas was initially linked to the webshop, resulting in orders to automatically be processed, however this linkage was terminated after the owner stopped cooperation with the designer of the webshop. Hence, the system now serves as a tool to perform administrative tasks and structures the order processing process. The owner recognizes that they are not making optimal use of the system due to both high work pressure and a lack of discipline. Although being used suboptimal, the system does allow the owner to have a clear overview of what is happening within his organization and it does structure the activities for which the system is being used. This results in higher efficiency and more transparency.

(25)

25 3.3 Data Analysis

The aim of the data analysis was to identify and explicate the critical events associated with the process of organizing of new ventures, specifically focussing on the role technology plays in this process. Secondly, relevant contextual factors and elements of the physical and social structure were identified to see how these combined to generate the emergent forces causal to the events identified in the previous step. Being an iterative process identifying these elements occurred simultaneously, although cases were analysed separately given the contextual differences. In order to derive a rich amount of data from the transcribed interviews and observations, a first analytical step was that of open coding. During open coding, data is broken down into discrete parts, closely examined and compared for similarities and differences (Strauss and Corbin, 1998). Interview transcripts and field notes were coded in Atlas.ti v7.5. An example of the open coding process is included in appendix G. During the open coding phase codes were primarily derived from the words the participants used themselves (In-vivo) or by using the vocabulary used in the previous chapters. In order to increase transparency after initial coding, codes with similar meanings were merged to prevent confusion. The final list included 41 codes, which all have been included in a codebook (See appendix H.).

After open coding, axial codingd was performed, per case, in order to group the different codes into separate categories. Axial coding involves coding for conditions (causal conditions, intervening conditions, and contextual conditions), actions and interactions, and consequences (Strauss and Corbin, 1998). In deriving this information from the data both the critical events related to this phenomenon of interest and the contextual and structural properties were extracted from the different cases.

Following axial coding, selective codinge was used. Selective coding is a process of integrating and refining categories with the goal of building and refining theory (Strauss and Corbin, 1998). During selective coding the axial codes derived from the various cases were integrated in order to distinguish commonalities amongst cases. Throughout this process the focus was both on the conditions and consequences of introducing the enterprise system within the new ventures. A primary sensitizing device for doing so was the model by Volkoff

et al. (2007), which allowed identification of the phases of structural conditioning, social

interaction and structural elaboration and reproduction. The theory concerning emergence helped with identifying patterns within the cases and contributed to the final step of

d

The results of axial coding process are included in appendix I to L.

(26)

26 retroducing generative mechanisms. In order to increase transparency of the retroduction process, memos were logged. An overview of the memos is included in Appendix Nf.

(27)

27

4. Retroduced Mechanisms

The goal of a critical realist study is to identify causes of the events of interest (Wynn and Williams, 2012). Moreover, to identify the generative mechanisms that are said to have caused the event under investigation. Mechanisms are “nothing other than the way of acting

of things” (Bhaskar, 1975, p. 14). Given that mechanisms are rarely, if ever, experienced

directly, a retroduced mechanism presents a logical argument explaining how the phenomenon of interest came to be through the emergent properties of the structure interacting within the study context (Wynn & Williams, 2012). In retroduction, “we take

some unexplained phenomenon and propose hypothetical mechanisms that, if they existed, would generate or cause that which is to be explained” (Mingers, 2004, p. 94). Put

differently, we seek to answer the question: “What properties must exist for [the phenomenon

of interest] to exist and to be what [it] is” (Danermark et al., 2002, p. 97).

Throughout data collection and analysis, by means of disassembling the data and repeated reading of the interviews and fieldnotes, “thought trials” were conducted to identify and describe the elements of the causal mechanisms, and the contextual influences responsible for its activation (Wynn and Williams, 2012). Backed by previous critical realist literature by Volkoff et al. (2007), mechanisms were theorized. Using the data collected, these mechanisms were confirmed through a process of emperical corraboration. This seeks to use data from observations and experiences to ensure that the proposed mechanisms adequately represent reality, and both have sufficient causal depth and better explanatory power than alternative explanations for the focal phenomenon (Wynn and Williams, 2012).

(28)

28 4.1 An embeddedness mechanism

The first mechanism identified is an embeddedness mechanism. Although the term embeddedness has been used in various ways, it is used here to describe the way in which technology introduces a material aspect to organizational elements such as routines, roles and data (Volkoff et al., 2007). The embeddedness mechanism emerged during users interaction with the technology. In two cases users are confronted with technology whose materiality had already been preconfigured for them (Leonardi, 2013), whilst in the remaining cases the users were themselve the designers of the software. Embeddedness emerges when the enterprise system is being used to perform certain routines, and the systems materiality mediates human agency or substitutes routines previously performed manually.

Interestingly, two of the companies within the data set did not acquire a preconfigured system from an external vendor, but decided to build a system of their own. [Company_A] chose to do so due to the fact that none of the systems of external vendors met up to their expectations. Moreover, the fact that [Owner_A.2] had former experience as a CRM developer provided another condition for making a system of their own.

It wasn’t totally clear how it would look like …. After a while we got to know that we, for example, needed multiple flexible packages, the contract, the invoice, all those things they could go better, go faster, more transparent …. That’s why we made a system of our own, because everything which was available in the market wasn’t good enough for us. [Owner_A.1, 14/12/2015]

In designing the system both owners delivered the input for what should be in the system and how it should look like. An external party did the necessary programming for making the system work. After launching the system the owners were able to work more efficiently and had a better overview of their internal processes, as is explained in appendix B.

It has become a series of steps that have to be followed: first you add the customer,

configure, sign the contract, and if all went well you were allowed to move on to installation. [Owner_A.1 14/12/2015]

(29)

29 the owners where themselves the designers of the system and embedded their (initial) ostensive knowledge of routines and data within the system.

When relating the above events to the model by Volkoff et al. (2007), the owners, by running in to various situations, gained an initial idea of how to perform certain routines and which options should be available. Due to frustrations about the length of this process the owners decided to build the system of their own, embedding the ostensive knowledge they had gained within the system. Once launched, the system started mediating their routines by requiring a certain input, or substituted routines previously executed manually. This phase of using the system and adopting its “logic” corresponds to the phase of social interaction. Moreover, it relates to the concept of supervenience, given the fact that the emerged order, by being embedded within the system, is now imposing structure on its underlying components (Sawyer, 2001). If the owners would run into unexpected (non-automated) circumstances the system would be adapted to meet the circumstances. This relates to the phase of structural elaboration and/or reproduction.

[Owner_A.1] argues that the system will facilitate further growth of the organization and will tell new employees exactly what to do once hired.

Someone who would start working here doesn’t need to do anything, everything is already there, and he doesn’t need to do anything regarding the administration. The CRM package is meant to enable my organization to grow, real structured growth. [Owner_A.1 14/12/2015]

In essence, [Company_A], by building the system themselves, moves through the entire model as was described by Volkoff et al. (2007). By making use of the system it has become imbricated within the organization. Sociomaterial practice, resulted in the structuring of the social subsystem, although the social subsystem was an important source for the systems initial materiality. The systems materiality has become interlocked within the organization’s structure, thus mediating the routines and data within the organization. It is through this interlocking the organizations structure was gradually formed.

(30)

30

From that point on the system started playing a bigger role…. We first had to process all the cases that needed attention before we could move on to other matters…. The system continuously showed popups, and you could only continue after you followed the rules.

[Owner_C.1 18/01/2016]

Moreover, [Owner_C.1] stated the following:

The way of working within our organization had become a bit more structured, because several regulations had been embedded, as fixed points, in the system.

The above statements are an example of how the system is mediating the sequence of routines performed by the users within the organization. Given the fact that the owners were obliged to adhere to the regulations, the systems materiality became further imbricated within the organization’s structure. In essence the system provoked a certain way of working from its owners, which resulted in sociomaterial practice affecting the social subsystem of the organization. This caused the system to become deeper entangled within the organization’s structure. In terms of Volkoff et al. (2007)’s model, structural elaboration occurred.

The same is apparent in [Company_B] and [Company_D], as is explained in corresponding appendices C and E. Both organizations encounter a system which sequenced steps influence the way in which the users perform their routines and which input is expected of them. In both organizations this resulted in higher efficiency. Moreover the way in which the system displayed the data resulted in its users having a better overview of the internal business processes.

In short, the embeddedness mechanism emerged in all cases within this data sample. Although operating in different contexts and having different backgrounds, the enterprise systems used within these organizations started mediating the routines performed within the organization after its introduction. This resulted in both structural elaboration and/or reproduction within these organizations.

(31)

31 4.2 A conditioning mechanism

The model by Volkoff et al. (2007) describes how the embedded organizational elements in the enterprise system mediate between the ostensive and performative aspects of the individuals using the system. However, this model assumes that the users of the system have a general understanding regarding the way they should perform certain routines. Given the fact that new ventures start out with no set of institutionalized routines yet, and enact their efforts outside established structures (Aldrich and Yang, 2013), the embedded organizational elements within the technology could serve as a direct source for the entrepreneurs to draw upon when organizing their venture.

In analysing the cases a conditioning mechanism emerged, wherein the embedded organizational elements formed a direct source for the entrepreneurs to draw upon for how they should execute certain routines. In this way the enterprise system serves as a sense giving device, and offers a template as how to organize and process certain data.

An example of how technology is a direct source for how to perform certain routines becomes apparent when the owner of [Company_D] decides to let go of his associate, who was responsible for the administrative tasks within the organization. Directly after [Owner_D.2] left, [Owner_D.1] acquired an enterprise system. This system not only automated certain processes, but also provided certain logic as how to perform the administrative tasks. The owner did not follow any training sessions from the vendor of the software given the fact that, at the time, these were deemed too expensive.

That’s when I let go of my associate, and I realized things had to go different…. that’s when I acquired Afas.

[Owner_D.1 21/01/2016]

Although initially acquiring the system to have a better overview of the internal business processes and the link between Afas and the webshop, this link only was of temporary nature. Appendix E gives a more elaborate explanation regarding these events. The system however remained operative, serving as an administrative tool.

This is however our central system, all invoices are send from this package, administrative tasks, can also be done in this package.

(32)

32 The enterprise system had become an essential part of the organization, and provided the logic of how to perform certain tasks. An example of this is the fact that the system required [Owner_D.1] to follow a preconfigured series of steps whenever processing an invoice, or performing other administrative tasks.

I can’t directly send an invoice from “sales order.” I first have to make a packing slip, confirm it, before I can move on to the next field, it is a sequence of steps.

[Owner_D.1 21/01/2016]

Despite the fact that the owner could have had a general understanding of how to perform the administrative routines previously performed by [Owner_D.2], and make up a logical series of steps of his own, the owner chose to copy the logic and templates proposed by the system, something commonly done by entrepreneurs when faced with the challenges of developing new routines (Aldrich and Yang, 2013). In this role, the system is not mediating between the ostensive and performative aspects of the routine, but forms a direct input for the ostensive knowledge surrounding the routine. Put differently, the systems structure conditions the ostensive knowledge of its users, thus a copying takes place between the ostensive as embedded by the designer, and the ostensive knowledge of the user.

This nuance also emerged in [Company_C], explained in more detail in appendix D.

Processing the job sheets into the system was a task that needed to be done, but it

didn’t have our main focus. It was a matter of quickly putting it in and moving on to more important matters…. our focus wasn’t on the administrative stuff.

[Owner_C.1 18/01/2016]

(33)

33 routine, which was copied by the owners of the organization. Thus, the systems embedded logic of how to deal with these situations was copied mindlessly by the owners of the organziation

Furthermore, although not applicable to the owners of [Company_A], the system served a same purpose.

Someone who would start working here doesn’t need to do anything, everything is already there, and he doesn’t need to do anything regarding the administration. The CRM package is meant to enable my organization to grow, real structured growth. [Owner_A.1 14/12/2015]

What [Owner_A.1] is actually stating here is the fact that a new employee, who has no frame of reference regarding their system, will be taught by the system how to perform certain routines. Essentially the user is required to insert certain data and follow certain logic. He is relieved of the struggle of having to develop the routines himself. In doing so, the system will leave an initial imprint within this user, adding to his field of reference.

Similary, in [Organization_B], the ES went live without instruction for the sales employees of how to use the system.

The system should explain itself, they provided the URL and the login …. see how it works. [Sales_1 14/01/2016]

Since [Company_B] already had a certain structure in place, the effects of introducing the new system resulted in minor changes to its structure. However, the fact that the system should be able to explain itself, builds the argument that the systems materiality could be copied by organizations or individuals in order to bridge a knowledge gap regarding how to perform certain routines, or how to structure certain data.

(34)

34 on their own. Hence a transfer occurred between the systems embedded material aspects and the organizations social subsystem. In doing so, an organizational structure emerged through the copying of these elements.

(35)

35

5. Theory building

As was argued in the beginning of this paper, new ventures, starting out with a blank slate and enacting their efforts outside established structures (Aldrich and Yang, 2013) initiate a process of organizing. In building an organization, entrepreneurs attempt to establish routines and order, but this effort is continuously disrupted by events, experiences, and interactions with other individuals that inject new information into the entrepreneurs’ cognitive framework (Morris and Webb, 2015). This combining of components leads to the creation of novel, resultant emergent states (McKelvey, 2004). In short, the entrepreneur has to create a new entity and in doing so can choose to develop routines himself by means of trial and error, or imitate and follow templates designed by others (Aldrich and Yang, 2013).

During data analysis, by means of retroduction, an embeddedness and conditioning mechanism emerged. These mechanisms both determined that the enterprise system used within these new ventures contain embedded organizational elements (e.g. routines, regulation and data), as that the ES, in the instance of being acquired from an external vendor, caused a certain conditioning to occur. Whilst the former has already been elaborated upon by in the model by Volkoff et al. (2007), the latter shows a new mechanism, which is particulary triggered by the fact that the new ventures start out with a blank slate. Regarding the latter, Gorman and Sandefur (2011) already pointed out that the embedding of what was previously “professional knowledge” gives the entrepreneur acces to routines with which they have no direct experience. Essentially this codification can be copied by the entrepreneur in order to surpass the dificulties of developing the routines, or templates on their own.

(36)

36

Proposition 1: Due to starting out with a blank slate, new ventures will copy the embedded “professional logic” of an enterprise system, when faced with the difficulties of having to design it on their own.

The enterprise system thus becomes yet another source from which entrepreneurs can draw upon when organizing the venture. In doing so the organizations structure is more directly affected by the materiality used by the organization, due to the conditioning mechanism that the system offers. This will enable the organization to stabilize faster and focus on its core business. Due to stabilization, effects as higher efficiency, more transparency and a reduced amount of errors occur. These effects were all present after introduction of the enterprise system within the new ventures in this data sample. Whilst two companies in this data set achieved this by embedding their own organizational elements within the system themselves, [Company_C] and [Company_D] copied certain “professional logic” due to events occurring during their organizing processes.

(37)

37

6. Discussion and Conclusions

Having retroduced the mechanisms through which technology causes the structuring of new ventures, it becomes interesting to see what scientific and managerial implications these finding might have. This chapter serves to present both.

6.1 Scientific and managerial implications

Previous literature on information systems primarily focussed on its effects in established structures. Although this helped develop an understanding of the effects technology has on these structures (e.g. Volkoff et al, 2007) it fell short of expanding their theories to a context wherein these structures were not yet in place. In this study a conditioning mechanism emerged which resulted in copying of the embedded organizational elements in to the new ventures structure. In doing so the entrepreneurs were relieved of having to develop certain routines on their own, causing structuring to occur faster.

Since this study is primarily explanatory, and focuses especially on the role of technology in this organizing process, it does not suggest the same mechanisms to be present within other contexts. Moreover, two of the organizations within this study decided to build a system of their own, in which the conditioning mechanism was less apparent.

However, the fact that this mechanism was present within two of the organizations in this study suggests that other new ventures could experience similar effects. Although copying these elements could constrain innovative activity heavily, in some situations this copying could help create more freedom for the entrepreneurs (Aldrich and Yang, 2013).

Further research should devote more attention to understanding the effects of enterprise systems in the organizing process of new ventures, especially since the choice for a certain system could constrain the perception of its users on the long term. This becomes apparent in the work by Leonardi (2013) where he recognizes the fact that technology could leave a certain imprint which is not easily changed. Moreover, path dependency could emerge due to the system becoming entangled within the organizations structure. One organization in this study already experienced the effects of a system malfunction. How much could the existence of an organization dependent on the system it uses.

(38)

38 6.2 Limitations

“Given the highly complex phenomena that are the focus of IS research, we cannot expect identical or even highly similar outcomes if we were to replicate a given study in a different organizational, industry, regional, or cultural setting” (Wynn and Williams, 2012,

p. 804).

Since each new venture follows its own unique emergent process in which all variables present influence the structure of the organization, the mechanism retroduced here do not necessarily have to cause similar effects in other settings. Research based on critical realism offers an explanation of how and why a mechanism causes a certain phenomena to occur in a specific setting, thus does not necessarily produce generalizable results.

Secondly, given the limited amount of time in which this study had to be conducted, data was primarily derived from the interviews conducted within the organizations. Although the observations did add valuable insights to how the entrepreneurs are mediated by the technology they did not help in explicating the conditioning mechanism since they did not capture the copying process.

Finally, the semi-structured interviews used in this study bears a researcher bias, thus decreasing the reliability and validity of the data collected. The same holds true regarding the retroduction of mechanisms. Although retroduced mechanisms were assessed according to criteria by Runde (1998), the quality of retroduction relies heavily on a researcher’s ability of doing so.

6.3 Conclusion

Most research in the field of IS primarily focuses on the effects of technology in established firms. New ventures however, by starting out with a blank slate and enacting their efforts outside established structures (Aldrich and Yang, 2013), could respond differently to the embedded organizational elements present in enterprise systems. Since entrepreneurs, when faced with the challenges of having to develop routines on their own, mostly follow and imitate templates designed by others (Aldrich and Yang, 2013), the enterprise system could form yet another source for the entrepreneurs to draw upon.

(39)

39 interviews, observations, and informal conversations). From the data, critical events, structures and contextual influences were explicated serving as an input for retroducing generative mechanisms. During retroduction, two generative mechanisms emerged playing a role in the organizing process of the new ventures in this study. An embedding mechanism emerged, wherein the enterprise systems used within the new ventures embedded certain organizational elements (e.g. routines, data and regulation), which caused mediation of the routines performed by the users of the system. Secondly, a conditioning mechanism emerged, which caused the entrepreneurs to copy the embedded “professional logic” present in the enterprise system. In doing so the entrepreneurs took shortcuts in their organizing processes by directly adopting the systems embedded organizational elements within the social subsystem of the organization. This caused the venture to stabilize faster, causing higher efficiency, more transparency and a reduced amount of errors.